Ectopic pregnancies represent 1–2% of all pregnancies (100,000 per year in the United States1 and 12,000 per year in the United Kingdom2) and are a leading cause of maternal morbidity and mortality in the first trimester, responsible for 3–8% of all pregnancy-related deaths,2,3 Ectopic pregnancies can be treated surgically or medically by administering intramuscular methotrexate. A systematic meta-analysis concluded the cost-effectiveness of medical treatment with methotrexate drops significantly with higher pretreatment human chorionic gonadotropin (hCG) levels (greater than 1,500 international units/L) and that laparoscopic excision remains the most effective treatment for ectopic pregnancy.3 Hence, many cases are treated surgically and there exists a need for more effective medical therapies to reduce operative intervention (and its inherent risks) in women diagnosed with ectopic pregnancy.
Epidermal growth factor (EGF) receptor signaling activates a potent cell survival response.4 Interestingly, placenta has, by far, the highest expression of EGF receptor compared with all other nonmalignant tissues in the body.5,6 Also, there is good evidence placental tissue heavily relies on EGF receptor signaling. Epidermal growth factor receptor signaling promotes cytotrophoblast motility,7 blocks apoptosis (programmed cell death),8 and rescues the placenta from apoptosis when exposed to hypoxia or toxic stimuli.9 Furthermore, genetic deletion of the EGF receptor in mice (EGF receptor knockout mice) results in midgestation lethality and gross placental abnormalities on histology.10 Therefore, inhibiting the EGF receptor could negatively affect placental survival.
Gefitinib is a molecularly targeted drug that selectively inhibits the tyrosine kinase domain of the EGF receptor. By inhibiting tyrosine kinase, gefitinib blocks EGF receptor signaling. Gefitinib has good oral bioavailability and is used clinically to treat breast cancer and non–small-cell lung cancers where it is taken indefinitely.11,12 Given placental survival and growth seems to be highly dependent on the EGF receptor pathway, it is plausible adding a short course of oral gefitinib to methotrexate could significantly improve its efficacy in resolving ectopic pregnancies.
Therefore, we performed preclinical studies to examine whether gefitinib enhances the ability of methotrexate to inhibit placental cell growth. We investigated the effects of adding gefitinib to methotrexate in three types of placental cells in vitro: JEG3 cells (choriocarcinoma cell line, models trophoblast cells), BeWo cells (choriocarcinoma cell line, models syncytiotrophoblast, the cell layer covering the surface of the placenta), and cells isolated from first-trimester placenta. We also investigated whether combining these agents was more potent in blocking the activation of EGF receptor and causing apoptosis than either agent alone. Lastly, we examined the effects of adding gefitinib to placental tissue in two animal models.
MATERIALS AND METHODS
We collected fallopian tube biopsies from the ectopic implantation site from participants (aged 18–45 years, n=5) undergoing surgery for an ectopic pregnancy at The Royal Infirmary of Edinburgh, Scotland. Biopsies (from ampullary region implantation sites) were collected immediately at operation and taken to the laboratory on ice for further processing. They were fixed in 10% neutral-buffered formalin overnight at 4°C, stored in 70% ethanol, and wax-embedded. Ethical approval was obtained from the Lothian Research Ethics Committee, Edinburgh, Scotland. Written informed consent was obtained from all participants.
Normal first-trimester tissue was collected from women undergoing elective termination of pregnancy at Monash Medical Centre, Clayton, Victoria, Australia. Samples were obtained by suction curettage and tissues washed in 0.9% saline and transferred into Dulbecco's modified eagle's media/F12 Ham media before transportation to the laboratory on ice. In the laboratory, primary cytotrophoblast cells were isolated as previously described13 and purity confirmed by immunocytochemistry for cytokeratin 7.14 This study of first-trimester placental tissue was approved by The Southern Health Human Research and Ethics Committee B (Monash Medical Centre), Victoria, Australia. All participants gave written, informed consent and were invited to participate only after the decision for a termination of pregnancy was made.
For many of our in vitro experiments, we used the placental cell lines JEG3 and BeWo cells purchased from the American Type Culture Collection. These are choriocarcinoma-derived cell lines widely used to study placental biology. JEG3 and BeWo cells were cultured in Dulbecco's modified eagle's media/F-12 medium containing L-glutamine supplemented with 10% fetal calf serum. BeWo cells were syncytialized with 100 micromolar of forskolin. Treatments were undertaken using methotrexate or gefitinib.
To perform immunohistochemistry of EGF receptor, paraffin sections of ectopic pregnancy specimens (3–4 micrometers thick) were dewaxed in xylene, rehydrated, and subjected to antigen retrieval by microwaving (700-W microwave) in sodium citrate (EGF receptor) for 20 minutes. Endogenous peroxidases the were blocked by application of 3% hydrogen peroxidase. An avidin-biotin block (for EGF receptor) and protein block then were applied for 10 minutes at room temperature. Sections were incubated overnight with EGF receptor mouse monoclonal antibody. For control sections, isotype control antibodies were applied. Staining was revealed by application of a biotinylated secondary horse antimouse antibody and ABC-Elite for EGF receptor. Positive immunostaining was visualized using 3,3-diaminobenxidine. Sections then were counterstained with Harris hematoxylin before being rehydrated and mounted with DPX mounting medium.
Cell viability after various treatments was assayed using the xCELLigence Real-Time Cell Analyzer SP instrument. The instrument was placed in a humidified incubator maintained at 37°C with 5% CO2. JEG3 cells were seeded at 625–40,000 cells per well in 96-well plates in medium containing 1% or 10% serum. Cells were initially monitored once every 2 minutes for 1 hour and then once every hour. After the addition of treatments, cells were monitored once every 10 minutes for 3 hours and thereafter once every hour. Cell viability was also analyzed in other experiments, as indicated, using the CellTiter-Blue Cell Viability Assay according to the manufacturer's instructions.
To perform Western blot analysis of EGF receptor, cells were trypsinized, washed and lysed in RIPA buffer (50 mM Tris, 150 mM NaCl, 1% triton, 0.1% sodium dodecyl sulphate, 1 mM ethylenediamine tetraacetic acid, 0.1% sodium deoxycholate, protease inhibitor cocktail, sodium vanadate) for 10 minutes before 15 minutes’ sonication, and centrifugation at 4°C to remove cellular debris. Lysates were resolved by reducing sodium dodecyl sulphate–polyacrylamide gel and transferred to polyvinyl difluoride membranes using the iBlot dry transfer system. Anti-pY845 EGF receptor rabbit polyclonal antibody was used to detect phosphorylated EGF receptor and anti-C-terminal EGF receptor rabbit polyclonal antibody 1005 was used to detect total EGF receptor. The same cell lysates were also analyzed for levels of phosphorylated Akt, p38 mitogen-activated protein kinase, IκB kinase, or extracellular signal-regulated kinase1/2 using the Bio-Plex Phosphoprotein Assays according to the manufacturer's instructions.
Apoptosis was measured with a one-step bioluminescent Caspase-Glo 3/7 Assay. Annexin V expression (a marker of apoptosis) was assessed by staining cells with a Phycoerythrin conjugated Annexin V antibody by fluorescence activated cell sorting as previous reported.15
To perform mouse xenograft experiments, we inoculated 5- to 6-week-old female SCID mice (C.B-17-Igh-1b-Prkdcscid; Animal Resources Centre, Perth, Australia) with JEG3 cells (106 in 100 mL phosphate-buffered saline) subcutaneously. Treatments started at day 7 after JEG3 injection with gefitinib (0.5–2 mg per dose in 50 mL dimethyl sulfoxide), methotrexate (0.04–0.4 mg per dose in 100 mL buffer), or a respective vehicle (dimethyl sulfoxide or buffer) per intraperitoneal injection. Xenograft volume in cubic millimeters was determined using the formula (length×width2)/2. We euthanized the mice after 19 days, at which time we excised the xenografts and collected blood by cardiac puncture. We quantified hCG in mouse serum using the Elegance hCG enzyme-linked immunosorbent assay following the manufacturer's instructions. For all xenograft experiments, four or more SCID mice were used per group.
To perform fetal resorption experiments, immunocompetent BalbC/Jamsu mice were time-mated and injected intraperitoneal at embryonic day 10 (E10) with gefitinib alone (100 mg/kg), methotrexate alone (20 mg/kg), gefitinib and methotrexate, or vehicle control. To control for methotrexate (carbonate buffer) and gefitinib vehicles (dimethyl sulfoxide), all mice were injected with the same amounts of both. Mice were culled at E15, uterine horns removed, and implantation sites counted as viable or resorbed. All animal studies were approved by the Monash University Animal Welfare Committee. There were five to seven pregnant BalbC/Jamsu mice and 30–43 implantation sites in each group.
We used the Student’s t test to compare two groups and a one-way analysis of variance with post hoc t tests to compare more than two groups. Fisher’s exact test was used to compare categorical variables.
Although normal placenta is known to strongly express EGF receptor,16 its presence in the placenta of ectopic pregnancies has yet to be confirmed. Therefore, we examined tubal ectopic pregnancy implantation sites (Fig. 1A) and verified EGF receptor is strongly expressed (Fig. 1B) on the syncytiotrophoblast of the placenta (the syncytiotrophoblast is a multinuclear layer covering the surface of the placenta and abuts the maternal circulation). As expected, EGF receptor is expressed specifically on the outer membrane of the syncytiotrophoblast, facing the maternal circulation. Epidermal growth factor receptor is also expressed in the underlying cytotrophoblast (Fig. 1B), cells that differentiate into the syncytiotrophoblast. We also confirmed EGF receptor is highly expressed in the placental cell types we used in subsequent experiments (JEG3 cells, BeWo cells, and first-trimester placenta; data not shown).
We then examined whether EGF receptor blockade adversely affects placental cell survival in vitro. We added gefitinib alone, methotrexate alone, and both drugs in combination to JEG3 cells in vitro. We examined growth curves in real time using the xCELLigence System. This is an assay that measures electrical impedance in real time where the presence of more cells causes greater impedance and a higher reading. This assay allows experiments to be monitored continuously in real time. In contrast, most existing cell viability assays only provide a reading at a single time point.
Methotrexate alone inhibited growth of JEG3 placental cells in a dose-dependent manner (Fig. 2A). Gefitinib alone had little effect on JEG3 placental cell growth (Fig. 2B). However, when gefitinib was added at increasing concentrations to a fixed dose of 100 micromolar of methotrexate, a potent additive effect was observed (Fig. 2C). For example, 100 micromolar of methotrexate combined with 4 micromolar of gefitinib had the same inhibitory effect on placental cell growth as 500 micromolar of methotrexate alone.
We verified these findings by adding these drugs to three types of placental cells and performing an end-point cell viability assay: JEG3 cells, syncytialized BeWo cells, and purified first-trimester trophoblast cells. In these experiments, we found adding both gefitinib and methotrexate was more potent in inhibiting growth of all three placental cell types compared with either agent alone (Appendix 1, available at http://links.lww.com/AOG/A417).
Next, we explored how gefitinib and methotrexate may be acting together on cell signaling pathways to inhibit placental cell growth. We administered gefitinib and methotrexate to JEG3 cells and examined how these agents affect EGF receptor phosphorylation (phosphorylation activates EGF receptor signaling, where adaptor molecules relay signals downstream). For these experiments, we also administered EGF as a stimulus to induce EGF receptor phosphorylation. As expected, gefitinib blocked EGF receptor phosphorylation (Fig. 2D). Interestingly, methotrexate alone decreased total EGF receptor and phosphorylated EGF receptor. Administering both agents was more potent in blocking EGF receptor phosphorylation than either alone. These data suggest gefitinib and methotrexate additively block EGF receptor signaling.
We also measured the phosphorylation status of four key signaling molecules downstream of the EGF receptor.17 Administering methotrexate alone was associated with an increase in Akt phosphorylation, an effect that was inhibited by gefitinib (Appendix 2, available online at http://links.lww.com/AOG/A418). A possible interpretation of this finding is that gefitinib may be blocking a compensatory survival response of Akt phosphorylation caused by single-agent methotrexate. For the three remaining downstream EGF receptor molecules examined (p38 mitogen-activated protein kinase, extracellular signal-regulated kinase1/2, and IκB kinase), there were no obvious trends to suggest these drugs were acting together on these proteins (data not shown).
We next investigated whether gefitinib and methotrexate were inducing apoptosis (programmed cell death) in JEG3 cells. Compared with either methotrexate or gefitinib alone, combination treatment significantly increased the apoptotic markers Annexin V (Annexin V detects the presence of phosphatidylserine on the outer membrane layer of cells, an event that is specific to apoptosis) and caspase 3 (caspase 3 is a protease that cleaves proteins within cells and is also activated during apoptosis; see Appendix 2, http://links.lww.com/AOG/A418). These data suggest these drugs are specifically inducing cell death, not only inhibiting cell growth.
Next, we undertook in vivo experiments in which we assessed the ability of methotrexate and gefitinib to inhibit subcutaneous JEG3 xenografts in SCID mice (immunodeficient mice used because they do not reject foreign tissue such as human placental cells). We developed this model because there are no established in vivo models of ectopic pregnancies, a uniquely human disease. We observed a dose-dependent decrease in xenograft volume with methotrexate treatment alone (Fig. 3A). In contrast to the in vitro cell viability assays where gefitinib alone caused little or no cell death, single-agent gefitinib in vivo also resulted in potent dose-dependent reductions in xenograft volume (see Fig. 3A and C; the experiment shown in Fig. 3C is different from Fig. 3B in that lower doses of gefitinib were used and the drug was administered twice, not three times).
We then examined whether combining the two drugs would result in an increased reduction in xenograft volume. To do this, we used a 0.04-mg/dose of methotrexate and a 0.5-mg/dose of gefitinib because these doses only caused a small reduction in xenograft volume as single agents (see Fig. 3B and C). Treatment with gefitinib and methotrexate potently decreased xenograft tumor volume compared with either agent alone (Fig. 3D). By day 19 after xenografting, mean (±standard error of the mean) xenograft volumes were: 821 (±68) mm3 after gefitinib treatment, 901 (±204) mm3 after methotrexate treatment, and 345 (±137) mm3 when both drugs were given at the same dose (P<.01, for both comparisons of single agents compared with the combination therapy).
We next measured serum hCG collected at the time of euthanasia from animals in the experiment depicted in Figure 3D. We did this because hCG is used clinically as a blood biomarker to assess the size of ectopic pregnancies. We found a significant decrease in serum hCG levels with combination treatment compared with single agents (Fig. 3E). We therefore conclude the combination of gefitinib and methotrexate can potently regress placental cell xenografts more effectively than either agent alone in SCID mice.
Given the xenograft model was undertaken in immunocompromised hosts, it was important to show gefitinib and methotrexate are able to regress placental tissue in an in vivo model where the host has an intact immune system. To do this, we examined the ability of these agents to cause fetal resorption (ie, fetal loss) within the uterus during early pregnancy in BalbC/Jamsu mice. We treated mice at embryonic day 10, euthanized them at embryonic day 15, and quantified the relative numbers of unaffected fetuses and those partially or fully resorbed. Gefitinib and methotrexate caused a rate of fetal resorption of 77%, approximately double the rates of resorption in mice treated with either agent or vehicle control (39–43% fetal resorption among these three groups; P≤.003 for all comparisons of single agents compared with combination treatment; see Appendix 3, available online at http://links.lww.com/AOG/A419). Appendix 3 also depicts examples of unaffected and resorbed embryos. We conclude the combination treatment is more effective in inducing fetal resorption in immunocompetent mice compared with either agent alone.
Epidermal growth factor receptor signaling triggers a series of intracellular pathways that result in cellular growth, proliferation, blockade of apoptosis signals, and stimulation of angiogenesis.12,17 Gefitinib is an orally available tyrosine kinase inhibitor that blocks EGF receptor signaling. Given the EGF receptor is very highly expressed in placental tissue, we hypothesized it would also be present in ectopic placental tissues and that gefitinib may be effective in treating ectopic pregnancy. We performed preclinical studies to examine this possibility.
We first demonstrated EGF receptor was highly expressed on the surface of placental tissues from ectopic pregnancies. Next we showed combining gefitinib with methotrexate potently induces placental cell death, confirming this observation in three placental cell types. We then examined molecular mechanisms and found both drugs may be acting in concert to block two key signaling molecules in the EGF receptor pathway, including the EGF receptor itself. We also found administering the drugs together increases placental apoptosis compared with either alone. Finally, we administered gefitinib and methotrexate systemically to two animal models. The drugs in combination potently decreased the volume of subcutaneous placental cell xenografts implanted in immunocompromised mice. To examine whether these agents affected placental tissue in immunocompetent mice, we administered gefitinib and methotrexate to pregnant mice with an intact immune system and found the drug combination doubled the rates of fetal resorption. Taken together, this body of preclinical work suggests the combination of gefitinib and methotrexate may be a promising approach to medically treat ectopic pregnancy.
Strengths of our in vitro work include the fact that we showed gefitinib and methotrexate were able to potently induce placental cell death in a number of placental cell types and that we were able to consistently demonstrate a dose-dependent effect. A limitation of our in vitro studies is that most of our experiments were done using placental cell lines (immortal placental cells derived from choriocarcinoma tissue) rather than primary placental cells. However, it was impractical for us to conduct all our experiments exclusively with primary cells isolated from first-trimester placenta (obtained from women undergoing termination of pregnancy) given they are difficult to obtain in the quantities required. Importantly, however, we confirmed our key observation in primary trophoblast cells (Supplementary Fig. 1C and D). We believe performing experiments in cell lines and validating key observations in primary tissues is an acceptable approach.
Our in vivo experiments were limited by the fact that neither animal model is particularly specific to ectopic pregnancy. We are not aware that an accepted animal model of ectopic pregnancy exists. The xenograft animal model, commonly used in oncology research to screen agents for antitumor activity,18 was limited by the fact nonpregnant immunocompromised hosts were used. Given this deficiency, we tested gefitinib and methotrexate in a second animal model in which the host was both immunocompetent and pregnant. However, in this model, the fetuses were within the uterus and not at an ectopic location. Nevertheless, we believe the findings from these in vivo experiments still support the premise that these drugs have potential merit in treating ectopic pregnancy because they show systemic administration of gefitinib and methotrexate can adversely affect placental tissue.
The clinical implication of our data is that adding gefitinib to the current protocol of single-agent methotrexate may be more effective than methotrexate alone in resolving ectopic pregnancies. A review by Jurkovic and Wilkinson19 suggests management of ectopic pregnancy using methotrexate is limited in clinical practice because only 25–30% of all ectopic pregnancies that present satisfy clinical criteria for medical management. It is therefore possible that combining gefitinib with methotrexate could result in a more efficacious treatment for ectopic pregnancy than methotrexate alone. If so, this drug combination could increase the proportion of ectopic pregnancies that could be safely treated medically to avoid surgery. It is important to note this treatment would not be suitable in situations in which there is a clinical suspicion the ectopic pregnancy has ruptured.
Given the encouraging results of these preclinical data, we have proceeded to translate these findings into the clinic. We have completed a phase I study in which we have administered combination gefitinib and methotrexate to women with ectopic pregnancies20 and we are actively recruiting participants for a phase II multicenter trial (ACTRN12611001056987).
1. Centers for Disease Control and Prevention (CDC). Ectopic pregnancy—United States 1990–1992. MMWR Morb Mortal Wkly Rep 1995;44:46–8.
2. Lewis GE. Saving Mothers’ Lives: reviewing maternal deaths to make motherhood safer—2003–2005. London (United Kingdom): The Confidential Enquiry into Maternal and Child Health (CEMACH); 2007.
3. Mol F, Mol BW, Ankum WM, van der Veen F, Hajenius PJ. Current evidence on surgery, systemic methotrexate and expectant management in the treatment of tubal ectopic pregnancy: a systematic review and meta-analysis. Hum Reprod Update 2008;14:309–19.
4. Herbst RS, Fukuoka M, Baselga J. Gefitinib—a novel targeted approach to treating cancer. Nat Rev Cancer 2004;4:956–65.
5. Su AI, Cooke MP, Ching KA, Hakak Y, Walker JR, Wiltshire T, et al.. Large-scale analysis of the human and mouse transcriptomes. Proc Natl Acad Sci U S A 2002;99:4465–70.
6. Wu C, Orozco C, Boyer J, Leglise M, Goodale J, Batalov S, et al.. BioGPS: an extensible and customizable portal for querying and organizing gene annotation resources. Genome Biol 2009;10:R130.
7. LaMarca HL, Dash PR, Vishnuthevan K, Harvey E, Sullivan DE, Morris CA, et al.. Epidermal growth factor-stimulated extravillous cytotrophoblast motility is mediated by the activation of PI3-K, Akt and both p38 and p42/44 mitogen-activated protein kinases. Hum Reprod 2008;23:1733–41.
8. Johnstone ED, Mackova M, Das S, Payne SG, Lowen B, Sibley CP, et al.. Multiple anti-apoptotic pathways stimulated by EGF in cytotrophoblasts. Placenta 2005;26:548–55.
9. Wolff GS, Chiang PJ, Smith SM, Romero R, Armant DR. Epidermal growth factor-like growth factors prevent apoptosis of alcohol-exposed human placental cytotrophoblast cells. Biol Reprod 2007;77:53–60.
10. Sibilia M, Wagner EF. Strain-dependent epithelial defects in mice lacking the EGF receptor. Science 1995;269:234–8.
11. Cohen MH, Williams GA, Sridhara R, Chen G, McGuinn WD Jr, Morse D, et al.. United States Food and Drug Administration Drug Approval summary: Gefitinib (ZD1839; Iressa) tablets. Clin Cancer Res 2004;10:1212–8.
12. Ciardiello F, Tortora G. EGFR antagonists in cancer treatment. N Engl J Med 2008;358:1160–74.
13. Cartwright JE, Kenny LC, Dash PR, Crocker IP, Aplin JD, Baker PN, et al.. Trophoblast invasion of spiral arteries: a novel in vitro model. Placenta 2002;23:232–5.
14. Paiva P, Salamonsen LA, Manuelpillai U, Walker C, Tapia A, Wallace EM, et al.. Interleukin-11 promotes migration, but not proliferation, of human trophoblast cells, implying a role in placentation. Endocrinology 2007;148:5566–72.
15. Kaitu'u-Lino TJ, Pattison S, Ye L, Tuohey L, Sluka P, MacDiarmid J, et al.. Targeted nanoparticle delivery of doxorubicin into placental tissues to treat ectopic pregnancies. Endocrinology 2013;154:911–9.
16. Ferretti C, Bruni L, Dangles-Marie V, Pecking AP, Bellet D. Molecular circuits shared by placental and cancer cells, and their implications in the proliferative, invasive and migratory capacities of trophoblasts. Hum Reprod Update 2007;13:121–41.
17. Dutta PR, Maity A. Cellular responses to EGFR inhibitors and their relevance to cancer therapy. Cancer Lett 2007;254:165–77.
18. Ciardiello F, Caputo R, Bianco R, Damiano V, Pomatico G, De Placido S, et al.. Antitumor effect and potentiation of cytotoxic drugs activity in human cancer cells by ZD-1839 (Iressa), an epidermal growth factor receptor-selective tyrosine kinase inhibitor. Clin Cancer Res 2000;6:2053–63.
19. Jurkovic D, Wilkinson H. Diagnosis and management of ectopic pregnancy. BMJ 2011;342:d3397.
20. Skubisz MM, Horne AW, Johns TG, Nilsson UW, Duncan WC, Wallace EM, et al.. Combination gefitinib and methotrexate compared with methotrexate alone to treat ectopic pregnancy. Obstet Gynecol 2013:122:745–51.